Project

Back to overview

Developing Capacity for High-Productivity Large-Scale Computing

English title Developing Capacity for High-Productivity Large-Scale Computing
Applicant Kuncak Viktor
Number 160453
Funding scheme SCOPES
Research institution Laboratoire d'analyse et de raisonnement automatisés EPFL - IC - IIF - LARA
Institution of higher education EPF Lausanne - EPFL
Main discipline Information Technology
Start/End 01.06.2015 - 30.06.2018
Approved amount 105'000.00
Show all

Keywords (6)

statistical computation; computer science MSc curriculum; Scala programming language; model-driven development; parallel computing; domain-specific languages

Lay Summary (French)

Lead
Le but de ce partenariat institutionnel SCOPES entre la Faculté Informatique et Communications de l'Ecole Polytechnique Fédérale de Lausanne (EPFL, Suisse) et la Faculté des Sciences de l'Université de Novi Sad (Serbie) est de développer la capacité de l'Université de Novi Sad à rechercher et enseigner dans le domaine du calcul à grande échelle. Ce domaine se réfère aussi bien aux calculs parallèles de haute performance, notamment utilisés lors de la résolution d'équation différentielles issues de la mécanique quantique, qu'au traitement statistique de larges volumes de données dans les domaines de l'intelligence artificielle, du data-mining ou encore de l'apprentissage machine qui peuvent, par exemple, être appliqués au processus de décision dans une entreprise.
Lay summary
Le but de ce partenariat institutionnel SCOPES entre la Faculté Informatique et Communications de l'Ecole Polytechnique Fédérale de Lausanne (EPFL, Suisse) et la Faculté des Sciences de l'Université de Novi Sad (Serbie) est de développer la capacité de l'Université de Novi Sad à rechercher et enseigner dans le domaine du calcul à grande échelle. Ce domaine se réfère aussi bien aux calculs parallèles de haute performance, notamment utilisés lors de la résolution d'équation différentielles issues de la mécanique quantique, qu'au traitement statistique de larges volumes de données dans les domaines de l'intelligence artificielle, du data-mining ou encore de l'apprentissage machine qui peuvent, par exemple, être appliqués au processus de décision dans une entreprise.
 
Direct link to Lay Summary Last update: 15.04.2015

Responsible applicant and co-applicants

Publications

Publication
Gaussian basis implementation of the charge patching method
Bodroski Zarko, Vukmirović Nenad, Skrbic Srdjan (2018), Gaussian basis implementation of the charge patching method, in Journal of Computational Physics, 368, 196-209.
Performing Hierarchical Clustering on Distance Matrices in OptiML
Fodor Lidija, Tesendic Danijela, Kurbalija Vladimir, Skrbic Srdjan (2017), Performing Hierarchical Clustering on Distance Matrices in OptiML, in Proceedings of the World Congress on Engineering, London, UK.
Proactive Synthesis of Recursive Tree-to-String Functions from Examples
Mayer Mikaël, Hamza Jad, Kuncak Viktor (2017), Proactive Synthesis of Recursive Tree-to-String Functions from Examples, in 31st European Conference on Object-Oriented Programming (ECOOP 2017), 74, Schloss Dagstuhl--Leibniz-Zentrum fuer Informatik, Schloss Dagstuhl, Germany 74.
Towards a Compiler for Reals
Darulova Eva, Kuncak Viktor (2017), Towards a Compiler for Reals, in ACM Trans. Program. Lang. Syst. (TOPLAS), 39(2), 1-28.
OpenMP, OpenMP/MPI, and CUDA/MPI C programs for solving the time-dependent dipolar Gross-Pitaevskii equation
Loncar Vladimir, Young-S Luis E., Skrbic Srdjan, Muruganandam Paulsamy, Adhikari Sadhan K., Balaz Antun (2016), OpenMP, OpenMP/MPI, and CUDA/MPI C programs for solving the time-dependent dipolar Gross-Pitaevskii equation, in COMPUTER PHYSICS COMMUNICATIONS, 209, 190-196.
CUDA programs for solving the time-dependent dipolar Gross-Pitaevskii equation in an anisotropic trap
Loncar Vladimir, Balaz Antun, Bogojevic Aleksandar, Skrbic Srdjan, Muruganandam Paulsamy, Adhikari K. Sadhan (2016), CUDA programs for solving the time-dependent dipolar Gross-Pitaevskii equation in an anisotropic trap, in Computer Physics Communications, 200, 406-410.
DSL for web application development
Boberic Krsticev Danijela, Tesendic Danijela, Jovic Milan, Bajic Zeljko (2016), DSL for web application development, in Proceedings of the 6th International Conference on Information Society and Technology, -, Kopaonik.
SMT-Based Checking of Predicate-Qualified Types for Scala
Schmid Georg Stefan, Kuncak Viktor (2016), SMT-Based Checking of Predicate-Qualified Types for Scala, in Scala Symposium, ACM, New York, NY.
Towards a Scala embedded information system description DSL
Bajic Zeljko, Jovic Milan (2016), Towards a Scala embedded information system description DSL.
A performance analysis of the R language and an assessment of the capabilities for its improvement
Fodor Lidija, Skrbic Srdjan (2015), A performance analysis of the R language and an assessment of the capabilities for its improvement, in Proceedings of the 5th International Conference on Information Society and Technology.
Towards a Scala embedded information system description DSL
Bajic Zeljko, Vidakovic Jovana, Skrbic Srdjan, Towards a Scala embedded information system description DSL, in Proceedings of the 7th Symposium on Computer Languages, Implementations and Tools (ICNAAM 2017), Thessaloniki, Greece.
Towards the High Performance Method for Large-Scale Electronic Structure Calculations
Bodroski Zarko, Vukmirovic Nenad, Skrbic Srdjan, Towards the High Performance Method for Large-Scale Electronic Structure Calculations, in Proceedings of the High Performance Computing and Applications Third International Conference.

Collaboration

Group / person Country
Types of collaboration
University of Novi Sad Serbien (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
7th Symposium on Computer Languages, Implementations and Tools, within ICNAAM (International Conference of Numerical Analysis and Applied Mathematics) Talk given at a conference Towards a Scala embedded information system description DSL 25.09.2017 Thessaloniki, Greece Škrbic Srdjan;
World Congress on Engineering 2017, International Conference of Data Mining and Knowledge Engineering Talk given at a conference Performing Hierarchical Clustering on Distance Matrices in OptiML 04.07.2017 London, Great Britain and Northern Ireland Škrbic Srdjan;
Scala Symposium 2016 Talk given at a conference Towards a Scala embedded information system description DSL 30.10.2016 Amsterdam, Netherlands Škrbic Srdjan; Kuncak Viktor;
5th International Conference on Information Society and Technology Talk given at a conference A performance analysis of the R language and an assessment of the capabilities for its improvement 08.03.2016 Kopaonik, Serbien Škrbic Srdjan;
6th International Conference on Information Society and Technology Talk given at a conference DSL for web application development 28.02.2016 Kopaonik, Serbien Škrbic Srdjan;
Third International Conference on High Performance Computing and Applications Talk given at a conference Towards the High Performance Method for Large-Scale Electronic Structure Calculations 26.07.2015 Shanghai, China Škrbic Srdjan;


Knowledge transfer events



Self-organised

Title Date Place
IC Research Day 2018 07.06.2018 EPFL, Switzerland
3RD ANNUAL WORKSHOP OF SCOPES PROJECT No IZ74Z0_160453 - UNIVERSITY-INDUSTRY COOPERATION DAY 16.03.2018 Novi Sad, Serbien

Abstract

The goal of the proposed institutional partnership SCOPES project between the School of Computer and Communication Sciences at the Swiss Federal Institute of Technology Lausanne (EPFL, Switzerland) and the Faculty of Science at the University of Novi Sad (Serbia) is to develop capacity at the University of Novi Sad for research and teaching focusing on the domain of large-scale computing. Under large scale computing we understand both high-performance parallel computation, as used, for example, when solving partial differential equations arising from quantum mechanics, as well as big data statistical computation, as used in data mining and machine learning, applied, for example, to decision making within an enterprise. By developing curriculum and research agenda in this area, we aim to educate the next generation of scientists and engineers that can solve difficult problems in science, engineering, and society much more efficiently. Joint research activities will be fostered through co-supervision of doctoral students. New courses to introduce functional programming in Scala, as well as domain-specific language technology will be created on the partner side following the success of the massive open online courses given by EPFL at Coursera. New courses will be developed on both sides to cover parallel computing at undergraduate and graduate level. New space created at the University of Novi Sad will be equipped to enable suitable instruction of computer science related courses. A compute node will be obtained to enable the development of parallel software for research purposes. Management structure and experience at EPFL will provide useful guidance for a new Section of Computer Science that will be created within the Department of Mathematics and Informatics, to provide an administrative framework for computer science activities and a degree of autonomy.
-